Modern technologies of authentication and authorization of access play a significant role in ensuring the protection of information in various practical applications. We consider the most convenient and used in modern mobile gadgets face authentication, ie when the primary information to provide access are certain features of biometric images of the user’s face. Most of the systems use intelligent processing of biometric images, in particular, artificial intelligence technology and deep learning. But at the same time, as always in cybersecurity, technologies for violating biometric authentication are being studied and researched. In particular, to date, the most common attack is substitution (spoofing), ie when attackers use pre-recorded biometric images to gain unauthorized access to critical information. For example, this could be a photo and/or video image of a person used to unlock their smartphone. Protection against such attacks is very difficult, because it involves the development and study of technologies for detecting signs of life. The most promising in this direction are artificial intelligence techniques, in particular, convolutional neural networks (CNN). This is the practical application of intelligent processing of biometric images and is studied in this article. We review various CNN settings and configurations and experimentally investigate their effect on the effectiveness of signs of life detection. For this purpose, success and failure indicators of the first and second kind are used, which are estimated by the values of cross entropy. These are reliable and reproducible indicators that characterize the effectiveness of protection against spoofing attacks on biometric authentication on the face. The world-famous TensorFlow and OpenCV libraries are used for field experiments, photos and videos of various users are used as source data, including Replay-Attack Database from Idiap Research Institute.

Convolutional Neural Networks to Protect Against Spoofing Attacks on Biometric Face Authentication

Artificial intelligence in physiological characteristics recognition for internet of things authentication

Biometric palm vein authentication becomes more popular in the industry. This paper proposes an algorithm for palm vein image selection. The main idea of this method is an approximation of biometric template image through discrete Markov process and applying the theory of conditional Markov process to develop the algorithm, which allows to fully utilize the statistic redundancy of the biometric template. Given method can be effectively used after performing a thresholding process on an image, using the dynamic threshold value. CASIA MS Palmprint V1 Database dataset was used to present the examples of algorithm implementation. Morphological dilation and Zhang-Suen algorithms were used to evaluate and compare the effectiveness of the proposed palm vein image thinning algorithm. The results showed that the proposed method makes it possible to compute an image that preserves the vein thickness and closely matches the real vein pattern and is effective when using a template matching algorithm based on perceptual hash functions.

Algorithm of Palm Vein Image Selection for Biometric Authentication Systems Based on Multidimensional Markov Chain

Today biometric identity authentication technologies are widespread. These systems are implemented not only in enterprises, controlled-access facilities, but also on smartphones of ordinary users and in online applications. The problem of choosing one of the authentication methods remains urgent. This paper provides a comparative analysis of existing systems and concludes that one of the most common and persistent methods is facial authentication system. The most powerful types of attacks on the biometric system are attacks on the database of biometric templates and attacks on sensors for obtaining biometric characteristics. Attacks on biometric sensors or spoofing attack is aimed at impersonating another person through fake biometric data. The paper deals with the possibility of special attacks on the biometric system of authentication by face image. A new method of detecting fake attacks (spoofing attacks) is proposed. The method is based on the use of an artificial convolutional neural network which was trained using a Replay-Attack Database from Idiap Research Institute. The obtained results show high efficiency of the proposed method of detecting spoofing attacks: the probability that an attack will be detected is 94.98%.

Biometric Authentication Using Convolutional Neural Networks

/pdf/biometric-images-attacks-detecting-using-convolutional-1y3jkeogss.pdf

Mykhaylo Bagmut

Papers

Biometric Authentication Using Convolutional Neural Networks

Convolutional Neural Networks to Protect Against Spoofing Attacks on Biometric Face Authentication

Biometric Images Attacks Detecting Using Convolutional Neural Networks.